Hugging Face's logo

qualcomm
/
DDRNet23-Slim

Image Segmentation
PyTorch
LiteRT
real_time
android
Model card Files
xet
Community

DDRNet23-Slim: Optimized for Mobile Deployment

Segment images or video by class in real-time on device

DDRNet23Slim is a machine learning model that segments an image into semantic classes, specifically designed for road-based scenes. It is designed for the application of self-driving cars.

This model is an implementation of DDRNet23-Slim found here.

This repository provides scripts to run DDRNet23-Slim on Qualcomm® devices. More details on model performance across various devices, can be found here.

Model Details

  • Model Type: Model_use_case.semantic_segmentation
  • Model Stats:
    • Model checkpoint: DDRNet23s_imagenet.pth
    • Inference latency: RealTime
    • Input resolution: 2048x1024
    • Number of output classes: 19
    • Number of parameters: 6.13M
    • Model size (float): 21.7 MB
    • Model size (w8a8): 6.11 MB
Model Precision Device Chipset Target Runtime Inference Time (ms) Peak Memory Range (MB) Primary Compute Unit Target Model
DDRNet23-Slim float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 112.853 ms 2 - 51 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 94.143 ms 24 - 82 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 49.803 ms 2 - 62 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 60.429 ms 24 - 85 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 42.019 ms 2 - 27 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 30.633 ms 24 - 43 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 24.882 ms 1 - 45 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 228.0 ms 2 - 51 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 38.017 ms 24 - 82 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float SA7255P ADP Qualcomm® SA7255P TFLITE 112.853 ms 2 - 51 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float SA7255P ADP Qualcomm® SA7255P QNN_DLC 94.143 ms 24 - 82 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 42.111 ms 2 - 22 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 30.769 ms 24 - 48 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float SA8295P ADP Qualcomm® SA8295P TFLITE 55.726 ms 2 - 56 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float SA8295P ADP Qualcomm® SA8295P QNN_DLC 42.221 ms 24 - 84 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 42.095 ms 2 - 25 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 30.445 ms 24 - 45 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float SA8775P ADP Qualcomm® SA8775P TFLITE 228.0 ms 2 - 51 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float SA8775P ADP Qualcomm® SA8775P QNN_DLC 38.017 ms 24 - 82 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 26.863 ms 2 - 58 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 20.796 ms 24 - 85 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 15.934 ms 31 - 86 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 21.271 ms 2 - 54 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 15.165 ms 16 - 86 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 12.659 ms 6 - 66 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 18.07 ms 1 - 54 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 11.021 ms 23 - 103 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 8.896 ms 30 - 127 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim float Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 31.416 ms 24 - 24 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim float Snapdragon X Elite CRD Snapdragon® X Elite ONNX 24.369 ms 24 - 24 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 TFLITE 178.671 ms 9 - 78 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 Dragonwing RB3 Gen 2 Vision Kit Qualcomm® QCS6490 ONNX 300.238 ms 194 - 214 MB CPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) TFLITE 99.305 ms 1 - 39 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 QCS8275 (Proxy) Qualcomm® QCS8275 (Proxy) QNN_DLC 141.319 ms 6 - 62 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) TFLITE 50.658 ms 1 - 51 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 QCS8450 (Proxy) Qualcomm® QCS8450 (Proxy) QNN_DLC 78.639 ms 6 - 70 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) TFLITE 50.573 ms 0 - 18 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) QNN_DLC 73.623 ms 6 - 27 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 QCS8550 (Proxy) Qualcomm® QCS8550 (Proxy) ONNX 76.899 ms 80 - 95 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) TFLITE 51.373 ms 1 - 40 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 QCS9075 (Proxy) Qualcomm® QCS9075 (Proxy) QNN_DLC 74.412 ms 6 - 61 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) TFLITE 339.544 ms 15 - 25 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 RB5 (Proxy) Qualcomm® QCS8250 (Proxy) ONNX 226.102 ms 194 - 205 MB CPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 SA7255P ADP Qualcomm® SA7255P TFLITE 99.305 ms 1 - 39 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 SA7255P ADP Qualcomm® SA7255P QNN_DLC 141.319 ms 6 - 62 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) TFLITE 50.831 ms 0 - 16 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 SA8255 (Proxy) Qualcomm® SA8255P (Proxy) QNN_DLC 73.595 ms 6 - 28 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 SA8295P ADP Qualcomm® SA8295P TFLITE 58.416 ms 1 - 43 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 SA8295P ADP Qualcomm® SA8295P QNN_DLC 82.917 ms 6 - 65 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) TFLITE 51.024 ms 0 - 22 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 SA8650 (Proxy) Qualcomm® SA8650P (Proxy) QNN_DLC 73.539 ms 6 - 24 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 SA8775P ADP Qualcomm® SA8775P TFLITE 51.373 ms 1 - 40 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 SA8775P ADP Qualcomm® SA8775P QNN_DLC 74.412 ms 6 - 61 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile TFLITE 38.274 ms 1 - 52 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile QNN_DLC 55.735 ms 6 - 72 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 Samsung Galaxy S24 Snapdragon® 8 Gen 3 Mobile ONNX 57.442 ms 90 - 144 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile TFLITE 37.586 ms 0 - 45 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile QNN_DLC 40.483 ms 6 - 70 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 Samsung Galaxy S25 Snapdragon® 8 Elite For Galaxy Mobile ONNX 41.455 ms 83 - 133 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile TFLITE 65.901 ms 6 - 48 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 Snapdragon 7 Gen 4 QRD Snapdragon® 7 Gen 4 Mobile ONNX 249.818 ms 192 - 208 MB CPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile TFLITE 23.3 ms 1 - 43 MB NPU DDRNet23-Slim.tflite
DDRNet23-Slim w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile QNN_DLC 46.922 ms 6 - 84 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 Snapdragon 8 Elite Gen 5 QRD Snapdragon® 8 Elite Gen5 Mobile ONNX 41.371 ms 89 - 140 MB NPU DDRNet23-Slim.onnx.zip
DDRNet23-Slim w8a8 Snapdragon X Elite CRD Snapdragon® X Elite QNN_DLC 76.964 ms 7 - 7 MB NPU DDRNet23-Slim.dlc
DDRNet23-Slim w8a8 Snapdragon X Elite CRD Snapdragon® X Elite ONNX 78.193 ms 131 - 131 MB NPU DDRNet23-Slim.onnx.zip

Installation

Install the package via pip:

pip install qai-hub-models

Configure Qualcomm® AI Hub Workbench to run this model on a cloud-hosted device

Sign-in to Qualcomm® AI Hub Workbench with your Qualcomm® ID. Once signed in navigate to Account -> Settings -> API Token.

With this API token, you can configure your client to run models on the cloud hosted devices.

qai-hub configure --api_token API_TOKEN

Navigate to docs for more information.

Demo off target

The package contains a simple end-to-end demo that downloads pre-trained weights and runs this model on a sample input.

python -m qai_hub_models.models.ddrnet23_slim.demo

The above demo runs a reference implementation of pre-processing, model inference, and post processing.

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.ddrnet23_slim.demo

Run model on a cloud-hosted device

In addition to the demo, you can also run the model on a cloud-hosted Qualcomm® device. This script does the following:

  • Performance check on-device on a cloud-hosted device
  • Downloads compiled assets that can be deployed on-device for Android.
  • Accuracy check between PyTorch and on-device outputs.
python -m qai_hub_models.models.ddrnet23_slim.export

How does this work?

This export script leverages Qualcomm® AI Hub to optimize, validate, and deploy this model on-device. Lets go through each step below in detail:

Step 1: Compile model for on-device deployment

To compile a PyTorch model for on-device deployment, we first trace the model in memory using the jit.trace and then call the submit_compile_job API.

import torch

import qai_hub as hub
from qai_hub_models.models.ddrnet23_slim import Model

# Load the model
torch_model = Model.from_pretrained()

# Device
device = hub.Device("Samsung Galaxy S25")

# Trace model
input_shape = torch_model.get_input_spec()
sample_inputs = torch_model.sample_inputs()

pt_model = torch.jit.trace(torch_model, [torch.tensor(data[0]) for _, data in sample_inputs.items()])

# Compile model on a specific device
compile_job = hub.submit_compile_job(
    model=pt_model,
    device=device,
    input_specs=torch_model.get_input_spec(),
)

# Get target model to run on-device
target_model = compile_job.get_target_model()

Step 2: Performance profiling on cloud-hosted device

After compiling models from step 1. Models can be profiled model on-device using the target_model. Note that this scripts runs the model on a device automatically provisioned in the cloud. Once the job is submitted, you can navigate to a provided job URL to view a variety of on-device performance metrics.

profile_job = hub.submit_profile_job(
    model=target_model,
    device=device,
)

Step 3: Verify on-device accuracy

To verify the accuracy of the model on-device, you can run on-device inference on sample input data on the same cloud hosted device.

input_data = torch_model.sample_inputs()
inference_job = hub.submit_inference_job(
    model=target_model,
    device=device,
    inputs=input_data,
)
    on_device_output = inference_job.download_output_data()

With the output of the model, you can compute like PSNR, relative errors or spot check the output with expected output.

Note: This on-device profiling and inference requires access to Qualcomm® AI Hub Workbench. Sign up for access.

Run demo on a cloud-hosted device

You can also run the demo on-device.

python -m qai_hub_models.models.ddrnet23_slim.demo --eval-mode on-device

NOTE: If you want running in a Jupyter Notebook or Google Colab like environment, please add the following to your cell (instead of the above).

%run -m qai_hub_models.models.ddrnet23_slim.demo -- --eval-mode on-device

Deploying compiled model to Android

The models can be deployed using multiple runtimes:

  • TensorFlow Lite (.tflite export): This tutorial provides a guide to deploy the .tflite model in an Android application.

  • QNN (.so export ): This sample app provides instructions on how to use the .so shared library in an Android application.

View on Qualcomm® AI Hub

Get more details on DDRNet23-Slim's performance across various devices here. Explore all available models on Qualcomm® AI Hub

License

  • The license for the original implementation of DDRNet23-Slim can be found here.
  • The license for the compiled assets for on-device deployment can be found here

References

Community

Downloads last month
353
Inference Providers NEW
Image Segmentation
This model isn't deployed by any Inference Provider. 🙋 Ask for provider support